首页/论文/2003.00497

arXiv 2003.004972020-03-01cs.LGstat.ML

Novelty-Prepared Few-Shot Classification

National Key Laboratory for Novel Software Technology,Nanjing University（南京大学计算机软件新技术国家重点实验室）

Chao Wang, Ruo-Ze Liu, Han-Jia Ye, Yang Yu

英文摘要：

Few-shot classification algorithms can alleviate the data scarceness issue, which is vital in many real-world problems, by adopting models pre-trained from abundant data in other domains. However, the pre-training process was commonly unaware of the future adaptation to other concept classes. We disclose that a classically fully trained feature extractor can leave little embedding space for unseen classes, which keeps the model from well-fitting the new classes. In this work, we propose to use a novelty-prepared loss function, called self-compacting softmax loss (SSL), for few-shot classification. The SSL can prevent the full occupancy of the embedding space. Thus the model is more prepared to learn new classes. In experiments on CUB-200-2011 and mini-ImageNet datasets, we show that SSL leads to significant improvement of the state-of-the-art performance. This work may shed some light on considering the model capacity for few-shot classification tasks.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。

阅读原文

阅读 PDF ↗

arXiv 页面 ↗HTML 原文 ↗

本文内容

英文摘要相关论文

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。

继续探索

相关论文

NoveltyRank: 一种用于人工智能研究概念新颖性估计的检索增强框架

NoveltyRank: A Retrieval-Augmented Framework for Conceptual Novelty Estimation in AI Research

Stanford University（斯坦福大学）

NoveltyRank通过结合语义表示学习与检索技术，提出了一种用于评估人工智能研究概念新颖性的框架，并展示了轻量模型在新颖性评估中的优越性能。

2026-01-03cs.CLcs.LG

Zero-Shot制胜：对抗鲁棒的新颖性检测

Killing it with Zero-Shot: Adversarially Robust Novelty Detection

Sharif University of Technology（谢里夫理工大学）；Okinawa Institute of Science and Technology（冲绳科学技术大学院大学）

本文提出结合最近邻算法与ImageNet预训练模型的鲁棒特征，以增强新颖性检测（ND）在对抗攻击下的性能和鲁棒性，并在多种基准测试中显著超越现有最先进方法。

2025-01-25cs.LG

CLAREL: Classification via retrieval loss for zero-shot learning

Element AI

We address the problem of learning fine-grained cross-modal representations. We propose an instance-based deep metric learning approach in joint visual and textual space. The key novelty of this paper is that it shows that using per-image semantic supervision leads to substantial improvement in zero-shot performance over using class-only supervision. On top of that, we provide a probabilistic justification for a metric rescaling approach that solves a very common problem in the generalized zero-shot learning setting, i.e., classifying test images from unseen classes as one of the classes seen …

2020-04-05cs.CVcs.LGstat.ML

Hierarchical Novelty Detection for Visual Object Recognition

University of Michigan（密歇根大学）；Korea Advanced Institute of Science and Technology（韩国科学技术院）；Google Brain（谷歌大脑）

Deep neural networks have achieved impressive success in large-scale visual object recognition tasks with a predefined set of classes. However, recognizing objects of novel classes unseen during training still remains challenging. The problem of detecting such novel classes has been addressed in the literature, but most prior works have focused on providing simple binary or regressive decisions, e.g., the output would be "known," "novel," or corresponding confidence intervals. In this paper, we study more informative novelty detection schemes based on a hierarchical classification framework. F…

2018-06-15cs.CVcs.LGstat.ML

FIGR: Few-shot Image Generation with Reptile

École Polytechnique de Montréal（蒙特利尔综合理工学院）；McGill University（麦吉尔大学）

Generative Adversarial Networks (GAN) boast impressive capacity to generate realistic images. However, like much of the field of deep learning, they require an inordinate amount of data to produce results, thereby limiting their usefulness in generating novelty. In the same vein, recent advances in meta-learning have opened the door to many few-shot learning applications. In the present work, we propose Few-shot Image Generation using Reptile (FIGR), a GAN meta-trained with Reptile. Our model successfully generates novel images on both MNIST and Omniglot with as little as 4 images from an unse…

2019-01-08cs.CVcs.LGstat.ML

基于预训练分割模型的零样本异常检测

Zero-Shot Anomaly Detection with Pre-trained Segmentation Models

Imperial College London（帝国理工学院）；Friedrich–Alexander University Erlangen–Nürnberg（埃尔朗根-纽伦堡弗里德里希-亚历山大大学）

本研究针对零样本异常检测问题，在WINCLIP框架基础上集成零样本分割模型与前景实例分割，无需外部数据，在VAND 2023零样本赛道获第三，VisA数据集上样本/像素级平均F1-max达81.5/24.2。

2023-06-15cs.CVcs.LG

更多相关论文收起更多论文

Deep Reinforcement Learning for Single-Shot Diagnosis and Adaptation in Damaged Robots

ABV-Indian Institute of Information Technology and Management, Gwalior（ABV印度信息技术与管理学院（瓜廖尔））

Robotics has proved to be an indispensable tool in many industrial as well as social applications, such as warehouse automation, manufacturing, disaster robotics, etc. In most of these scenarios, damage to the agent while accomplishing mission-critical tasks can result in failure. To enable robotic adaptation in such situations, the agent needs to adopt policies which are robust to a diverse set of damages and must do so with minimum computational complexity. We thus propose a damage aware control architecture which diagnoses the damage prior to gait selection while also incorporating domain r…

2019-10-02cs.AIcs.LGcs.RO

A Review of Open-World Learning and Steps Toward Open-World Learning Without Labels

University of Colorado Colorado Springs（科罗拉多大学科罗拉多斯普林斯分校）

In open-world learning, an agent starts with a set of known classes, detects, and manages things that it does not know, and learns them over time from a non-stationary stream of data. Open-world learning is related to but also distinct from a multitude of other learning problems and this paper briefly analyzes the key differences between a wide range of problems including incremental learning, generalized novelty discovery, and generalized zero-shot learning. This paper formalizes various open-world learning problems including open-world learning without labels. These open-world problems can b…

2022-01-03cs.AIcs.CVcs.LG

L2AE-D：基于元级Dropout的少样本学习嵌入聚合方法

L2AE-D: Learning to Aggregate Embeddings for Few-shot Learning with Meta-level Dropout

University of Nottingham（诺丁汉大学）；Automated Scheduling Optimisation and Planning (ASAP) Research Group（自动调度优化与规划（ASAP）研究组）

本文提出L2AE-D方法，通过通道注意力聚合嵌入并引入元级Dropout，缓解少样本学习中的元级过拟合，在Omniglot和miniImageNet上达到最先进性能。

2019-04-08cs.CVcs.LG

自适应少样本学习用于两光子光刻中鲁棒部件质量分类

Adaptive few-shot learning for robust part quality classification in two-photon lithography

Department of Mechanical Engineering, University of Michigan, Ann Arbor, MI 48109, United States（机械工程系，密歇根大学，安娜堡，MI 48109，美国）

本文提出了一种自适应CV框架，通过统计假设检验、两阶段增量学习和域对抗神经网络，实现两光子光刻中部件质量分类的鲁棒性和数据效率。

2026-01-13cs.CVcs.LG

OpenPatch：用于分布外检测的三维拼布方法

OpenPatch: a 3D patchwork for Out-Of-Distribution detection

Politecnico di Torino（都灵理工大学）；Italian Institute of Technology（意大利技术研究院）

针对三维点云语义新类检测，本文提出无需训练的OpenPatch，从预训练模型中间特征提取已知类拼块表示，并依据新样本由单一或多类拼块重构的程度计算异常分数；在合成已知、真实测试场景及少样本设置中表现优异，且跨骨干与预训练目标稳健。

2023-10-23cs.CVcs.LG

MAML-KT：通过少样本模型无关元学习解决知识追踪中的冷启动问题

MAML-KT: Addressing Cold Start Problem in Knowledge Tracing for New Students via Few-Shot Model-Agnostic Meta Learning

New Mexico State University（新墨西哥州立大学）

本文提出MAML-KT模型，通过少样本元学习方法提升新学生知识追踪的早期准确性，实验显示其在ASSIST2009、ASSIST2015和ASSIST2017数据集上表现优于传统KT模型。

2026-04-14cs.AIcs.LG

Large-Scale Long-Tailed Recognition in an Open World

The Chinese University of Hong Kong（香港中文大学）；UC Berkeley（加州大学伯克利分校）；ICSI（国际计算机科学研究所）

Real world data often have a long-tailed and open-ended distribution. A practical recognition system must classify among majority and minority classes, generalize from a few known instances, and acknowledge novelty upon a never seen instance. We define Open Long-Tailed Recognition (OLTR) as learning from such naturally distributed data and optimizing the classification accuracy over a balanced test set which include head, tail, and open classes. OLTR must handle imbalanced classification, few-shot learning, and open-set recognition in one integrated algorithm, whereas existing classification a…

2019-04-16cs.CVcs.LG

Open Long-Tailed Recognition in a Dynamic World

Nanyang Technological University（南洋理工大学）；University of California, Berkeley（加州大学伯克利分校）；International Computer Science Institute（国际计算机科学研究所）；The Chinese University of Hong Kong（香港中文大学）；Google Inc.（谷歌公司）

Real world data often exhibits a long-tailed and open-ended (with unseen classes) distribution. A practical recognition system must balance between majority (head) and minority (tail) classes, generalize across the distribution, and acknowledge novelty upon the instances of unseen classes (open classes). We define Open Long-Tailed Recognition++ (OLTR++) as learning from such naturally distributed data and optimizing for the classification accuracy over a balanced test set which includes both known and open classes. OLTR++ handles imbalanced classification, few-shot learning, open-set recogniti…

2022-08-17cs.CVcs.LG

Linear Disentangled Representation Learning for Facial Actions

Johns Hopkins University（约翰斯·霍普金斯大学）

Limited annotated data available for the recognition of facial expression and action units embarrasses the training of deep networks, which can learn disentangled invariant features. However, a linear model with just several parameters normally is not demanding in terms of training data. In this paper, we propose an elegant linear model to untangle confounding factors in challenging realistic multichannel signals such as 2D face videos. The simple yet powerful model does not rely on huge training data and is natural for recognizing facial actions without explicitly disentangling the identity. …

2017-01-11cs.AIcs.CVcs.LG

利用传染模型原型预训练流行病时间序列预测器

Pre-training Epidemic Time Series Forecasters with Compartmental Prototypes

Emory University Department of Computer Science Atlanta GA USA（埃默里大学计算机科学系）；Emory University Rollins School of Public Health Atlanta GA USA（埃默里大学罗林斯公共卫生学院）；Emory University（埃默里大学）

CAPE 是首个开源的流行病预测预训练模型，通过预训练学习流行病学语义，提升预测鲁棒性和泛化能力。

2026-02-23cs.LG